Topaz-Denoise: general deep denoising models for cryoEM and cryoET

By Tristan Bepler, Kotaro Kelley, A. Noble, Bonnie Berger. 2019

523 citations recorded by the source corpus.

Cryo-electron microscopy (cryoEM) is becoming the preferred method for resolving protein structures. Low signal-to-noise ratio (SNR) in cryoEM images reduces the confidence and throughput of structure determination during several steps of data processing, resulting in impediments such as missing particle orientations. Denoising cryoEM images can not only improve downstream analysis but also accelerate the time-consuming data collection process by allowing lower electron dose micrographs to be used for analysis. Here, we present Topaz-Denoise, a deep learning method for reliably and rapidly increasing the SNR of cryoEM images and cryoET tomograms. By training on a dataset composed of thousands of micrographs collected across a wide range of imaging conditions, we are able to learn models capturing the complexity of the cryoEM image formation process. The general model we present is able to denoise new datasets without additional training. Denoising with this model improves micrograph interpretability and allows us to solve 3D single particle structures of clustered protocadherin, an elongated particle with previously elusive views. We then show that low dose collection, enabled by Topaz-Denoise, improves downstream analysis in addition to reducing data collection time. We also present a general 3D denoising model for cryoET. Topaz-Denoise and pre-trained general models are now included in Topaz. We expect that Topaz-Denoise will be of broad utility to the cryoEM community for improving micrograph and tomogram interpretability and accelerating analysis. The low signal-to-noise ratio (SNR) in cryoEM images can make the first steps in cryoEM structure determination challenging, particularly for non-globular and small proteins. Here, the authors present Topaz-Denoise, a deep learning based method for micrograph denoising that significantly increases the SNR of cryoEM images and cryoET tomograms, which helps to accelerate the cryoEM pipeline.

Subfields: Molecular Biology, Structural Biology, Surfaces, Coatings and Films

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Molecular Biology

In Structural Biology

In Surfaces, Coatings and Films

Citations in the corpus

EPicker is an exemplar-based continual learning approach for knowledge accumulation in cryoEM particle picking (2022, methodology): “Later, more particle picking approaches and programs, such as DeepEM 9 , Warp 10 , Topaz 11,12 , and crYOLO 13 , used CNN or modi ﬁ ed CNN.”
Complete mapping of mutations to the SARS-CoV-2 spike receptor-binding domain that escape antibody recognition (2020, unknown): “The images then were picked with Topaz (Bepler et al., 2019, 2020) .”
Isotropic reconstruction for electron tomography with deep learning (2022, methodology): “Deep learning was also applied to cryoET tomograms for denoising 27 – 29 , particle picking 30,31 , classi ﬁ cation 32,33 , and segmentation 34,35 , accelerating the interpretation and analysis of the tomograms.”
Structural insights into the assembly and activation of the IL‐27 signaling complex (2022, unknown): “Deep learning models used in Topaz (v.0.2.5; Bepler et al , 2020) were used to automatically pick particles from these micrographs within the cryoSPARC workflow.”
New tools for automated cryo-EM single-particle analysis in RELION-4.0 (2021, methodology): “For all data sets, except for the ribosome data set collected with a phase plate (ribo-VPP; EMPIAR-10153), particle picking using the pre-trained model in Topaz yielded reasonable results.”